Ex Parte Dombrowski et al

Patent Trial and Appeal Board

Apr 23, 2014

12027316 (P.T.A.B. Apr. 23, 2014)

UNITED STATES PATENT AND TRADEMARK OFFICE
____________

BEFORE THE PATENT TRIAL AND APPEAL BOARD

____________

Ex parte BERNHARD DOMBROWSKI, KARL KLUG,
MICHAL SKUBACZ, PETER SUDA, JURGEN TOTZKE,
and CAI-NICOLAS ZIEGLER
____________

Appeal 2011-013087
Application 12/027,316
Technology Center 2100
____________

Before JASON V. MORGAN, LARRY J. HUME, and
LINZY T. McCARTNEY, Administrative Patent Judges.

McCARTNEY, Administrative Patent Judge.

DECISION ON APPEAL
Appellants appeal under 35 U.S.C. § 134(a) the Examiner’s decision
finally rejecting claims 1-22. We heard oral argument on this appeal on
March 18, 2014. We have jurisdiction over the appeal under 35 U.S.C.
§ 6(b).
We AFFIRM-IN-PART and enter a NEW GROUND OF
REJECTION.

Appeal 2011-013087
Application 12/027,316

2
STATEMENT OF THE CASE
Appellants’ invention concerns “methods and apparatuses for
tracking or monitoring changes in web documents.” (Spec. 1.) Claim 1
illustrates the claimed subject matter:
1. A method for tracking a change in a collection of web
documents, said web documents being provided by a web site,
the method comprising:
retrieving the web documents of said collection of web
documents at a first assigned point in time and at a second
assigned point in time that is later than the first assigned point
in time;
calculating a similarity measure for a combination of a
first retrieved web document of the retrieved web documents at
the first assigned point in time and a second retrieved web
document of the retrieved web documents at the second
assigned point in time to determine whether the second
retrieved document corresponds with the first retrieved
document; and
if the second retrieved web document corresponds to the
first retrieved web document, comparing said calculated
similarity measure for said first and second retrieved web
documents with an upper similarity threshold for the similarity
measure for detecting a substantial change in content of the
corresponding first and second retrieved web documents.
REJECTIONS AND PRIOR ART
Claims 16-20 and 22 stand rejected under 35 U.S.C. § 112 ¶ 1 as
failing to comply with the written description requirement.
Claims 1-22 stand rejected under 35 U.S.C. § 103(a) as unpatentable
over Liang (US 2006/0047649 Al; March 2, 2006) in view of Canright (US
2007/0185871 Al; August 9, 2007).
Appeal 2011-013087
Application 12/027,316

3
CLAIM GROUPING
Based on Appellants’ arguments, we decide this appeal based upon
the following five groups of claims:
Group I: Claims 1-15 and 211
Group II: Claims 6-8
Group III: Claim 14
Group IV: Claims 16-18
Group V: Claims 19, 20, and 22.
(See App. Br. at 14-21.) Groups I, II, IV, and V stand or fall with claims 1,
6, 16, and 19, respectively, as Appellants have not presented separate
patentability arguments for the other claims in each group.
ISSUES
Appellants’ Appeal Brief raises the following five issues:2
(1) Did the Examiner erroneously find claims 16-20 and 22 fail to
satisfy the written description requirement of 35 U.S.C. § 112 ¶ 1?

1 Appellants did not address claim 21 in their Appeal Brief. Because claim
21 depends from claim 1, the rejection of claim 21 under 35 U.S.C. § 103(a)
stands or falls with claim 1.
2 Appellants’ Reply Brief includes arguments not presented in Appellants’
Appeal Brief. (See, e.g., Reply Br. at 3-4.) Appellants have waived these
arguments by failing to raise them in their Appeal Brief. See Ex parte
Borden, 93 USPQ2d 1473, 1474 (informative) (“[T]he reply brief [is not] an
opportunity to make arguments that could have been made in the principal
brief on appeal to rebut the Examiner’s rejections . . . .”). See also Becton
Dickinson & Co. v. C.R. Bard, Inc., 922 F.2d 792, 800 (Fed. Cir. 1990)
(“[A]n issue not raised by an appellant in its opening brief . . . is waived.”).
Accordingly, we consider only arguments Appellants properly raised in their
Appeal Brief.
Appeal 2011-013087
Application 12/027,316

4
(2) Did the Examiner erroneously find the combination of Liang and
Canright teaches or suggests the limitations recited in claim 1?
(3) Did the Examiner erroneously find the combination of Liang and
Canright teaches or suggests the limitations recited in claim 6?
(4) Did the Examiner erroneously find the combination of Liang and
Canright teaches or suggests the limitations recited in claim 14?
(5) Did the Examiner erroneously find the combination of Liang and
Canright teaches or suggests the limitations recited in claims 16 and 19?
ANALYSIS
Written Description Rejection
The Examiner finds Appellants’ specification does not adequately
describe “calculating similarity score values for comparing the first content
objects with the second content objects” as recited in claims 16 and 19. The
Examiner finds the specification only discloses comparing similarity score
values with thresholds and concludes these comparisons do not provide
sufficient written description support for the disputed limitation. (Ans. at 4-
5, 19-21.) In response, Appellants contend at least some of these
comparisons, when properly understood, provide sufficient support for the
limitation. (Reply Br. at 1-2.)
We agree with Appellants. Although the Examiner correctly notes
Appellants’ specification describes comparing similarity score values with
thresholds, the Examiner ignores the purpose of at least some of those
comparisons: determining whether first and second content objects
“correspond.” For example, Appellants’ specification discloses comparing
first and second content objects’ calculated similarity score value with a
lower threshold. (Spec. 8-9, 11.) If the value is above the threshold,
Appeal 2011-013087
Application 12/027,316

5
Appellants’ disclosed method assumes the objects correspond and processes
the objects accordingly. (Id. at 8-9.) This method uses the similarity score
value to “compare” content objects in that the method relies on the value to
determine whether the objects are sufficiently similar. We find Appellants’
disclosure provides adequate written description support for the disputed
limitation; therefore, we do not sustain the written description rejection of
claims 16-20 and 22.
Obviousness Rejection
Claim 1
The Examiner finds the combination of Liang and Canright teaches or
suggests the method recited in claim 1. The Examiner finds Liang discloses
a webpage monitoring process that (1) retrieves a webpage at specified
intervals, (2) calculates a checksum for the webpage, and (3) compares the
checksum with a previously calculated checksum for the webpage to
determine if the webpage has changed. (Ans. at 5-6, 22-23.) The Examiner
also finds Liang does not teach the recited “similarity measure” but finds
Canright discloses “similarity scores” that satisfy this element. (Id. at 6.)
The Examiner concludes the combination of Liang’s webpage monitoring
process and Canright’s similarity scores render claim 1 obvious. (Id. at 7.)
Appellants contend the cited references do not teach or suggest the
following two limitations recited in claim 1:
(1) “calculating a similarity measure . . . to determine whether
the second retrieved document corresponds with the first
retrieved document” and
(2) “if the second retrieved web document corresponds to the
first retrieved web document, comparing said calculated
similarity measure . . . with an upper similarity threshold for
Appeal 2011-013087
Application 12/027,316

6
the similarity measure for detecting a substantial change in
content.”
(See, e.g., App. Br. at 14.) Appellants argue neither reference teaches
comparing previously saved web documents, nor detects whether a newly
stored document from the same website has substantially changed from a
previously stored web document. (See id. at 14-15.)
We agree with the Examiner. Appellants essentially argue neither
Liang nor Canright individually teaches or suggests the recited limitations.
“But one cannot show non-obviousness by attacking references individually
where, as here, the rejections are based on combinations of references.” In
re Keller, 642 F.2d 413, 426 (CCPA 1981). As described above, the
Examiner finds the combination of Canright’s similarity score and Liang’s
webpage monitoring process suggests “calculating a similarity measure . . .
to determine whether the second retrieved document corresponds with the
first retrieved document,” and Appellants’ arguments have not persuaded us
the Examiner erred in this regard.
To the extent Appellants’ arguments rest on a belief the limitation
“determine whether the second retrieved document corresponds with the first
retrieved document” means determining which second document best
“corresponds” with a first web document, we disagree. This limitation
simply requires determining whether a first and second object “correspond,”
and we agree with the Examiner the proposed Liang-Canright combination
satisfies this limitation. We also agree the Examiner provided ample reason
to combine the cited art in the suggested manner.
As for the limitation beginning “if the second retrieved web document
corresponds,” the use of the conditional word “if” makes this limitation
Appeal 2011-013087
Application 12/027,316

7
optional; therefore, the Examiner need not find this limitation in the prior
art. See Ex Parte Katz, 2010-006083, 2011 WL 514314, at *4-5 (BPAI
2011) (non-precedential) (concluding an examiner was not required to find
an “if” limitation in the prior art when, under the broadest scenario, the
method need not invoke the steps).
Claim 6
The Examiner also concludes the combination of Liang and Canright
suggests “a similarity score value . . . exceeding a predetermined lower
similarity threshold value and falling below the upper similarity threshold”
as recited in claim 6. (Ans. at 9-10.) The Examiner finds Canright teaches
collecting document similarity scores above a similarity threshold and
setting the remaining scores to zero. (Id. at 24.) According to the Examiner,
this suggests similarity score values falling between a lower threshold (zero)
and an upper threshold (the aforementioned similarity threshold). (Id.) In
response, Appellants argue that Canright suggests only a lower threshold.
(Reply Br. at 6-7.)
We agree with Appellants for two reasons. First, the cited portion of
Canright discloses one threshold, not two as recited in claim 6. (See Ans. at
23-24 (citing Canright ¶¶ 0051, 0052, and 0063).) Canright does not suggest
zero is a threshold as found by the Examiner. Rather, Canright teaches
setting certain similarity matrix elements to zero when a document similarity
score falls below a similarity threshold. See Canright ¶ 52. Second,
Canright teaches the disclosed similarity threshold is a lower threshold, not
an upper threshold. Canright describes using the similarity threshold to
avoid calculating similarity scores for documents lying lower in a document
list. Id. ¶¶ 0051, 0052, and 0063. The cited passages do not suggest
Appeal 2011-013087
Application 12/027,316

8
stopping or otherwise altering the similarity value calculation for similarity
scores above the threshold.
Claim 14
The Examiner finds the combination of Liang and Canright also
teaches or suggests the limitations recited in claim 14. In particular, the
Examiner finds this combination suggests determining both “which content
objects of the first retrieved web document corresponds to content objects of
the second retrieved web document” and “detecting whether changes of
content in the content objects . . . exist.” The Examiner finds Canright
teaches calculating a “word similarity score” for the documents in which a
word appears. (Ans. at 24-25 (citing Canright, Figure 3, ¶ 0056).) (A word
similarity score equals a word’s normalized word frequency for a document.
See Canright, ¶ 0056.) The Examiner finds the disclosed words serve as the
recited “content objects” and finds the combination of Canright’s words and
word similarity scores and Liang’s webpage monitoring process suggests the
disputed limitations. (Ans. at 24-25.)
Appellants contend neither Liang nor Canright teaches the recited
limitations. (See, e.g., Reply Br. at 7-8.) Appellants contend words are not
“content objects,” and regardless, neither reference suggests identifying
corresponding content objects, much less detecting changes in content
objects. (See App. Br. at 18-19; Reply Br. at 7-8.)
We again agree with the Examiner. Here, too, Appellants argue the
references separately, and such arguments are unpersuasive for the reasons
discussed above, see Keller, 642 F.2d at 426. Appellants’ contention that
words cannot serve as “content objects” is equally unpersuasive.
Appellants’ specification explains that “content objects” are “web document
Appeal 2011-013087
Application 12/027,316

9
parts,” which include “text portions” and “text objects.” (See, e.g., Spec 4,
6.) Individual words are “text portions” and “text objects.” And like claim
1, claim 14 does not require determining which second object best or most
corresponds with a first object. Rather, claim 14 simply recites “determining
which content objects of the first retrieved web document corresponds [sic]
to content objects of the second retrieved web document.”
Claims 16 and 19
The Examiner finds the proposed Liang-Canright combination teaches
or suggests “determining which second content object corresponds to which
first content object by determining which second content object has a highest
similarity score value as corresponding to the first content object” as recited
in claim 16. The Examiner also finds this combination suggests a similar
limitation in claim 19. For both limitations, the Examiner notes Canright
teaches a similarity matrix in which the rows are organized in descending
order of each document’s similarity to an index word. (Ans. at 26-27.) The
Examiner finds this teaching suggests limitations at issue. (Id.)
Appellants argue, among other things, the cited art fails to teach or
suggest using a similarity score value to compare content objects retrieved at
different times. (See, e.g., App. Br. at 20.)
We conclude the Examiner erred. Claims 16 and 19 each claim
determining which content second content object corresponds to a first
content object by “determining which second content object has a highest
similarity score value” with respect to a first content object. The cited
paragraph of Canright simply teaches ranking documents in descending
order of “similarity” to a word; the paragraph is silent about using a highest
Appeal 2011-013087
Application 12/027,316

10
similarity score to determine which second content object corresponds to a
first content object.
New Ground of Rejection
Pursuant to our authority under 37 C.F.R. § 41.50(b), we enter a new
ground of rejection under 35 U.S.C. § 112 ¶ 4 for claims 21 and 22 as not
further limiting the claims from which they respectively depend.
Claims 21 and 22 are “computer program product” claims that depend
from method claims 1 and 16, respectively. However, claims 21 and 22 fail
to specify a further limitation of the subject matter of claims 1 and 16 to
which they respectively refer because claims 21 and 22 are completely
outside the scope of claims 1 and 16. Pfizer, Inc. v. Ranbaxy Labs. Ltd., 457
F.3d 1284, 1292 (Fed. Cir. 2006).
DECISION
We AFFIRM the rejections of claims 1-5, 9-15, and 21 under
35 U.S.C. § 103(a). We REVERSE the rejections of claims 16-20 and 22
under 35 U.S.C. § 112 ¶ 1. We REVERSE the rejections of claims 6-8, 16-
20, and 22 under 35 U.S.C. § 103(a). We enter a new ground of rejection for
claims 21 and 22 under 35 U.S.C. § 112 ¶ 4.
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a). See 37 C.F.R.
§ 1.136(a)(1)(iv).

AFFIRMED-IN-PART
37 C.F.R. § 41.50(b)

msc